SÃO PAULO STATE UNIVERSITY “JÚLIO DE MESQUITA FILHO” SCHOOL OF AGRICULTURAL AND VETERINARY SCIENCES JABOTICABAL CAMPUS UTILIZATION OF RECONSTITUTED CORN GRAIN SILAGE IN SUPPLEMENTS FOR FINISHING BEEF CATTLE ON PASTURE: EVALUATION OF SUPPLEMENT AEROBIC STABILITY AND CATTLE PERFORMANCE Saulo Teixeira Rodrigues de Almeida Animal Scientist 2024 SÃO PAULO STATE UNIVERSITY “JÚLIO DE MESQUITA FILHO” SCHOOL OF AGRICULTURAL AND VETERINARY SCIENCES JABOTICABAL CAMPUS UTILIZATION OF RECONSTITUTED CORN GRAIN SILAGE IN SUPPLEMENTS FOR FINISHING BEEF CATTLE ON PASTURE: EVALUATION OF SUPPLEMENT AEROBIC STABILITY AND CATTLE PERFORMANCE Master’s candidate: Saulo Teixeira Rodrigues de Almeida Advisor: Prof. Dr. Gustavo Rezende Siqueira Dissertation presented to the School of Agricultural and Veterinary Sciences – Unesp, Jaboticabal Campus, as part of the requirements for obtaining the Master’s degree in Animal Science. 2024 AUTHOR’S CURRICULAR DATA Saulo Teixeira Rodrigues de Almeida was born on June 8, 1996, in Belo Horizonte, MG, Brazil. He is the son of Laura Stella Teixeira and José Carlos Rodrigues de Almeida. He graduated in Animal Science from the Federal University of Lavras (UFLA, 2015 – 2021). During his undergrad, he was a member of the Brazilian Forage Team (NEFOR) (2017 – 2021), teaching assistant of “Analytical Chemistry” and “Biostatistics” courses and received a scientific initiation scholarship (FAPEMIG). He spent one year (2019 -2020) at The University of Maine, USA, completing an internship in feed conservation and ruminant nutrition. Saulo was awarded as the best student in the Animal Science Program. In 2022, he began his master’s at Unesp Jaboticabal and APTA Colina and received a FAPESP scholarship. In 2023, he spent 5 months at The University of Adelaide, Australia, as a visiting researcher, conducting research on barley processing and degradability. A minha mãe Laura, que nunca mediu esforços para fazer o que foi melhor para seu filho e sempre me apoiou incontestavelmente. A minha madrinha Sônia que também sempre esteve ao meu lado e ao de minha mãe em todos os momentos. Aos meu avô José de Almeida (in memorian) e minha avó Lilita (in memorian) que não se esqueceram de mim e tornaram toda essa caminhada possível. DEDICO! AGRADECIMENTOS Agradeço a Deus por ter me permitido concluir mais essa etapa da minha carreira e ter dado força para superar todos os desafios e amparo nos momentos difíceis. A minha mãe Laura e a minha tia Sonia por todo amor, apoio e incentivo. A minha namorada Isabela, por todo seu amor, sua companhia e paciência e por sempre estar me esperando com um sorriso no rosto. Ao meu orientador, Prof. Gustavo Siqueira, pela oportunidade de ser seu orientado, pelas contribuições nessa conquista, pelos ensinamentos e por me instigar a ter uma visão crítica, prática e direta. Ao Prof. Flávio pelos concelhos por abrir as portas da APTA para a universidade. Aos amigos do GEPROR, Aline, Amanda, Anna Lídia, Ariane, Fernanda e Fernanda Scherem, Gabriel, Gilyard, Iam, Igor, Iorrano, José Maria, Luciana, Luisão, Mateus e José Maria pela parceria e por todos bons momentos. Em especial agradeço ao Mateus, Igor, Iorrano e Gilyard por todos os debates e ensinamentos a respeito da nossa pecuária. E a Amanda pela ajuda em todos os plaqueamentos. A Prof. Mariana Caetano por ter me recebido e orientado durante o intercâmbio na Austrália. Ainda aos meus orientadores da graduação, Professores Thiago, Márcio e Juan, por terem despertado em mim o interesse pela pesquisa e por terem proporcionado o que foi preciso para que eu chegasse aonde estou. A todos os funcionários da APTA pelas ajudas durante as conduções dos experimentos e manejos, Miltinho, Rogério, Verde, Toinzinho, Lore, Sr Chico, Sueli, Néia, Regina, Toga, Antônio Carlos, Juninho, Roberto e Luizinho. Ao grupo LPCD pelo financiamento e pela confiança depositada em mim para condução do experimento. Ao Antônio, Marechal, Iago e Sr Luciano pelo apoio na condução do experimento e pelos ensinamentos. A Alltech e 3r Lab pela realização das análises e a Nutricorp pela doação do inoculante. A FAPESP (# processo 2022/08742-5) e CNPQ (# processo 130049/2022-0) pela concessão da bolsa. Muito obrigado! i SUMARY ANIMAL WELFARE CERTIFICATION ........................................................................ ii ABSTRACT ................................................................................................................ iii RESUMO.................................................................................................................... iv CHARPTER 1 – GENERAL CONSIDERATION .......................................................... 1 1. INTRODUCTION ..................................................................................................... 1 2. LITERATURE REVIEW ........................................................................................... 2 2.1 Corn grain physiology ......................................................................................... 2 2.2 Corn use and processing methods ..................................................................... 5 2.2.1 Whole corn and dry processing methods ........................................................ 5 2.2.2 Ensiled corn ..................................................................................................... 7 2.2.2.1 Reconstituted corn grain silage .................................................................... 9 2.2.3 Corn grain silage inoculants .......................................................................... 10 2.2.4 Beef cattle finishing phase on pasture with high supplementation ................ 13 3. OBJECTIVE AND HYPOTHESIS .......................................................................... 15 4. REFERENCES ...................................................................................................... 15 CHARPTER 2 - EXPERIMENT 1: EVALUATION OF SILAGE INOCULANTS FOR RECONSTITUTED CORN GRAIN SILAGE AND SUPPLEMENT AEROBIC STABILITY. ............................................................................................................... 23 1. INTRODUCTION .................................................................................................. 23 2. MATERIAL AND METHODS ................................................................................. 24 3. RESULTS .............................................................................................................. 29 4. DISCUSSION ........................................................................................................ 42 5. CONCLUSION ...................................................................................................... 45 6. REFERENCES ...................................................................................................... 46 CHARPTER 3 - EXPERIMENT 2: INTAKE AND PERFORMANCE OF FINISHING BEEF CATTLE ON PASTURE RECEIVING RECONSTITUTED CORN GRAIN SILAGE SUPPLEMENTATION ON PASTURE EXPOSED FOR DIFFERENT DAYS .................................................................................................................................. 49 1. INTRODUCTION .................................................................................................. 49 2. MATERIAL AND METHODS ................................................................................. 50 3. RESULTS .............................................................................................................. 55 4. DISCUSSION ........................................................................................................ 61 5. CONCLUSION ...................................................................................................... 64 6. REFERENCES ...................................................................................................... 64 ii iii RECONSTITUTED CORN GRAIN SILAGE UTILIZATION IN SUPPLEMENTS FOR FINISHING BEEF CATLE ON PASTURE: EVALUATION OF SUPPLEMENT AEROBIC STABILITY AND CATTLE PERFORMANCE ABSTRACT – Two experiments were conducted aiming to: 1) evaluate the combination of Lactobacillus buchneri + Lactobacillus hilgardii and Lactobacillus buchneri + Lactobacillus plantarum in the ensiling of reconstituted corn grains and their effects on silage fermentation, nutritional value, and aerobic stability of the silage and the supplement made with these silages; 2) to determine the impacts of providing fresh and exposed reconstituted corn grain silage supplements on the intake and performance of pasture-finished Nellore cattle with high supplementation. Experiment one evaluated reconstituted corn grain silages non-inoculated and inoculated with L. buchneri + L. hilgardii both in concentration of 1.5 x105 cfu/g and L. buchneri + L. plantarum at 2 x105 cfu/g and 1 x105 cfu/g respectively, stored for 60 days and supplements for beef cattle using these silages. Inoculated silages presented a higher number of lactic acid bacteria, pH and aerobic stability (P < 0.01) and lower number of yeast (P < 0.01) in comparison to non-inoculated and did not differ between both inoculated (P > 0.01). Treated supplements had higher aerobic stability and lower yeast and mold counts than control supplement and did not differ between them. In experiment 2, 70% of dry rolled corn was substituted to reconstituted corn grain silage. Cattle consuming reconstituted corn grain silage supplement until 6 days of aerobic exposure had higher average daily gain and final body weight (P ≤ 0.06) tended to consume less supplement (P ≤ 0.09) and were more efficient in putting weight per kg of dry matter ingested (P ≤ 0.04). Aerobic exposition had no linear effect and did not impact cattle performance in comparison to fresh supplement. Inoculation with L. buchneri + L. hilgardii or L. buchneri + L. plantarum guarantee aerobic stability for 9 days in silage and for 6 days in supplement under laboratory conditions. Cattle consuming reconstituted corn grain silage supplement exposed until 6 days had better performance than cattle consuming dry corn supplement. Keywords: finishing beef cattle, high supplementation on pasture, reconstituted corn grain silage iv UTILIZAÇÃO DE SILAGEM DE GRÃOS DE MILHO RECONSTITUÍDOS EM SUPLEMENTOS PARA BOVINOS DE CORTE EM TERMINAÇÃO A PASTO: AVALIAÇÃO DA ESTABILIDADE AERÓBICA DO SUPLEMENTO E DO DESEMPENHO DOS BOVINOS RESUMO – Foram conduzidos dois experimentos com o objetivo de: 1) avaliar a combinação de Lactobacillus buchneri + Lactobacillus hilgardii e Lactobacillus buchneri + Lactobacillus plantarum na ensilagem de grãos de milho reconstituídos e seus efeitos na fermentação, valor nutricional e estabilidade aeróbica da silagem e do suplemento feito com essas silagens; 2) determinar os impactos da oferta de suplementos de grãos de milho reconstituídos frescos e expostos sobre o consumo e desempenho de bovinos Nelore em terminação a pasto com alta suplementação. O experimento um avaliou silagens de grãos de milho reconstituídos não inoculadas e inoculadas com L. buchneri + L. hilgardii ambas na concentração de 1,5 x 105 ufc/g e L. buchneri (2 x 105 ufc/g) + L. plantarum (1 x 105 ufc/g) armazenadas por 60 dias, e suplementos para bovinos utilizando essas silagens. As silagens inoculadas apresentaram maior número de bactérias ácido lácticas, pH e estabilidade aeróbia (P < 0,01) e menor número de leveduras (P < 0,01) em comparação a silagem não inoculada, não diferindo entre silagens inoculadas (P > 0,01). O suplemento com silagem inoculada apresentou maior estabilidade aeróbia e menor contagem de leveduras e fungos do que o suplemento controle, e não diferiu entre eles. No experimento 2, 70% de milho seco moído foi substituído por silagem de grãos de milho reconstituídos. Os bovinos consumindo suplemento de silagem de grãos de milho reconstituídos por até 6 dias de exposição aeróbia apresentaram maior ganho médio diário e peso corporal final (P ≤ 0,05), tenderam a consumir menos suplemento (P ≤ 0,09) e foram mais eficientes em ganhar peso por kg de matéria seca ingerida. A exposição aeróbica não teve efeito linear e não impactou o desempenho dos bovinos em comparação com o suplemento fresco. A inoculação com L. buchneri + L. hilgardii ou L. buchneri + L. plantarum garante estabilidade aeróbica da silagem por pelo menos 9 dias e por 6 dias no suplemento em condições de laboratório. Os bovinos consumindo suplemento de silagem de grãos de milho reconstituídos expostos por até 6 dias tiveram melhor desempenho do que os bovinos consumindo suplemento com milho moído. Palavras-chave: bovinos de corte, alta suplementação a pasto, silagem de grãos de milho reconstituídos 1 CHAPTER 1 – GENERAL CONSIDERATIONS 1. Introduction Brazil was the second biggest bovine meat producer with 10.8 million tons of equivalent carcass in 2022 and the first in exportation (3.02 million tons). To achieve this number, 42.3 million animals were slaughtered and 81.8% came from pasture yielding 297 kg of carcass on average (ABIEC, 2023). Tropical pastures rarely provide the necessary nutrients to achieve high gains and fat deposition in bovines in the fattening phase (REIS et al., 2011). Feedlot fattening during this phase emerges as an alternative for producers to attain satisfactory results in weight gain in a short time, however, properties with feedlot facilities are scarce and the high investment necessary to build them makes it unfeasible for many producers. Thus, fattening with high supplementation on pasture becomes a viable option. The main energy resource used in Brazil to fatten beef cattle is corn. In the last 2 years, corn prices have varied a lot, and the exchange ratio of carcass:corn sack hit 3.42 sacks in 2022 (CEPEA, 2024). An alternative to protect against these low rates periods is improving the nutrient availability to the animal. Reconstituted corn grain silage (RCGS) increases starch degradability, thereby improving animal feed efficiency (FERRARETTO; CRUMP; SHAVER, 2013; HOFFMAN et al., 2011). In a general way, supplements utilized in the finishing phase are prepared in advance to be provided throughout the following days. The use of RCGS in this type of supplement is a challenge due to the high susceptibility of aerobic spoilage in comparison with the dry ingredients traditionally used. Thus, to use RCGS in supplements, it is necessary to guarantee the aerobic stability (AS) of the supplement during the whole time before use, avoiding losses in sanitary and nutritional quality, and, consequently, reducing of animal consumption and performance (HOFFMAN; OCHER, 1997; WHITLOCK et al., 2000). One way to guarantee AS is by inoculating obligatory heterofermentative bacteria during ensilage. This type of bacteria will produce mainly weak acids (e.g., acetic and propionic acids) that have antifungal action, inhibiting yeasts and molds (OUDE ELFERINK et al., 2001). Among the species in this group, L. buchneri is 2 recognized as an extensively studied bacteria and established as an effective silage inoculant, while L. hilgardii, though recently investigated as a silage inoculant and showing promising results, has not yet been fully established.(DA SILVA et al., 2021; DROUIN; TREMBLAY; CHAUCHEYRAS-DURAND, 2019). With this, there is the possibility of using wet ingredients to guarantee a more efficient and profitable supplementation strategy in high-supplementation grass-fed cattle finishing systems. 2. Literature review 2.1. Corn grain physiology The high grain yield and energy content of corn grain are among the primary reasons it is widely used in animal nutrition. Corn grain can be divided into three main parts: the pericarp (5~7% of the grain) is a fibrous layer covering the grain; germen (11~12% of the grain) is the protein and fat proportion of the grain responsible for starting germination; and the endosperm (80~85% of the grain) responsible for accumulating the grain energy reserves that will be used in the germination process (GARCÍA-LARA; CHUCK-HERNANDEZ; SERNA-SALDIVAR, 2019) (Image 1). Figure 1: Basic composition of corn grain. Available on: https://www.milkpoint.com.br/colunas/thiago-fernandes-bernardes/conhecendo-e- escolhendo-hibridos-de-milho-para-silagem-80791n.aspx?acao=8f27852b-6ad2-4fc8-a673- 01c24aae0ac8 Corn grain stores energy in starch form (70% of dry matter (DM) content). The main varieties of corn planted worldwide for animal nutrition propose are classified as flint or dent corn, according to the percentage of vitreous texture endosperm. Dent corn Endosperm Pericarp Germen 3 has a predominantly white and opaque floury endosperm (51.8% - 80%) (CORREA et al., 2002; ZHANG; XU, 2019) which means most of the starch granules are disposed as smaller spheres with a discontinuous and disorganized protein matrix in the endosperm interspace (GAYRAL et al., 2016; ZHANG; GAO; DONG, 2011). Flint corn has an organized and compacted protein matrix that covers and strongly links with larger polygonal starch granules in the majority part of the endosperm (~73%) (ZHANG; GAO; DONG, 2011) conferring a yellow and translucent apparence and making the grain harder and denser, hindering the access of water and microbes to the starch granules (Image 2). A greater proportion of vitreous endosperm also leads to higher electricity use during the processing of the grain. The vitreous endosperm has a higher concentration of amylose than the floury endosperm (ZHANG; XU, 2019) which is associated with higher retrogradation after the gelatinization process. The flat organization of glucose in amylose allows it to go into amylopectin forming hydrogen bonds within the starch molecule impacting enzyme activity and ability to expand and this may make it impossible to return to the original structure after retrogradation, thus decreasing degradability (ROONEY; PFLUGFELDER, 1986). Figure 2: Scanning electron micrograph of fracture yellow dent corn endosperm, showing round (A), thigh-packed starch granules (B), cell walls (Cw), and starch granules (Sg) (x1500). Available at García-Lara et al., (2019). Flint corn has been predominantly used in Brazil (CORREA et al., 2002) due to better adaptability to tropical weather and higher resistance to insect attack during the storage period. The organizational structure of starch granules in vitreous endosperm makes more difficult its degradation by microorganisms in the rumen. The predominant 4 protein (~75% among starch granules) in the corn endosperm is called zein, which is part of the prolamin group, followed by glutelin (~14%) (GAYRAL et al., 2016). Philippeau et al., (2000) described the vitreous endosperm as starch granules surrounded by protein storage bodies (zeins) in a protein matrix (glutelin) embedded in a dense matrix of endosperm cells (Image 3). The better organization of starch granules with protein filled in the interspace gives more density to flint grains (PEREIRA et al., 2008). Zein is divided in α, β, γ and δ subclasses. The vitreous endosperm has a higher concentration of protein than floury endosperm, principally due to a higher concentration of α-zein. (α, β, δ)-Zeins are positively correlated with grain vitreousness and negatively correlated with ruminal degradability (PEREIRA et al., 2008; PHILIPPEAU; LANDRY; MICHALET-DOREAU, 2000). When the zein barrier is covering starch granules, rumen microorganisms will have to hydrolyze the prolamin first to then have access to the starch. This process costs time and depending on the grain processing degree, starch will not be available and will go towards omasum. PHILIPPEAU; LANDRY; MICHALET-DOREAU, (2000) found greater concentration of rumen slowly degradable starch in flint corn when compared to dent corn (80.7 vs 73.4% of total starch). CORREA et al., (2002); PEREIRA et al., (2004) observed that starch degradability in vitreous corn decreased as grain maturity increased even after the black line point, directly affecting the nutritional value of the grain, probably because of increasing cross-linking between zeins. With advancing maturity, β and α zeins cross-link, and γ and δ zeins penetrate their net encapsulating starch granules into a hydrophobic starch-protein matrix (BUCHANAN; GRUISSEM; JONES, 2000; HOFFMAN et al., 2011). 5 Figure 3 Transmission electron microscopy showing starch granule (S) and protein body (PB) in dent corn (D) and popcorn (L). Available in Zhang et al., (2011). 2.2. Corn use and processing methods The main objective of processing grain is to increase nutritional value. When corn is processed, the pericarp is damaged, particle size is reduced, and starch surface area increases, facilitating microbial and enzymatic access to the starch granules, bosting digestibility. The effects of processing are more noticeable in grains with lower degradability (e.g., sorghum and flint corn) (JACOVACI et al., 2021; OWENS; BASALAN, 2013). 2.2.1. Whole corn and dry processing methods Whole corn is the simplest way to feed corn to ruminants. Usually, it is cheaper than any other corn processing since no energy is required to process the grain, but it requires that the animal chews the grain to damage its pericarp. Conflict in literature has been found about the advantage of using whole corn in beef cattle nutrition. GOROCICA-BUENFIL; LOERCH, (2005) fed whole and ground dent corn to cattle and got similar average daily gain (ADG), hot carcass weight (HCW), and dry matter intake (DMI) however more fecal starch were found for whole corn-fed steers. According to 6 the same author, 91.04% of the whole ingested kernels disappeared and 86.01% of the initial starch in a kernel was recovered in non-damaged excreted kernels. FREITAS et al., (2021), in contradiction with the previously mentioned experiment, found better ADG, HCW, lower DMI, and similar feed efficiency for animals eating ground corn diets in comparison with whole corn diets. OWENS; SODERLUND, (2006) in a compilation of data, concluded that the ruminal disappearance of whole corn grain starch is 68.34%, while dry-rolled corn starch is 63.8%. However, when total tract starch disappearance is analyzed, this value reverses to 91.03% for starch coming from dry rolled corn and 87.08% for whole corn grain starch. The lower ruminal starch degradability of rolled corn may be explained by the faster rate of passage of small particles, which can be accelerated when a diet contains high levels of physically effective fiber. Ground and rolled corn are the most common processing methods adopted in Brazil due to their facility (SILVESTRE; MILLEN, 2021) and to enhance the surface area exposed to microbial and enzymatic attack. Rolling corn consists of breaking the kernel using two rolls while grinding normally uses a hammer mill with a sieve. Rolled grains are characterized to have more uniform and larger particle size than ground grains (CORONA et al., 2005; NIKKHAH, 2012). No difference in ADG, feed efficiency (FE), and rumen pH was reported when ground corn was compared to dry rolled corn, still, starch digestion (86.3% vs 92%) and excretion (25% vs 16% of DM) were better for ground corn due to its smaller particle size (CORONA et al., 2005). SCHWANDT et al., (2016) observed a quadratic tendency for ADG and FE during the acclimatization time when cattle were fed medium-rolled corn (3.76 um) but no differences were seen when the entire feedlot period was analyzed, even with a linear increase in fecal starch as particle size increases. It is important to mention that the previous experiments were performed using dent corn. GOUVÊA et al., (2019) evaluating two different geometrical particle sizes of flint corn noticed a tendency of better ADG, final body weight, and HCW when corn was coarsely ground (2.12mm vs 1.66mm) without differences in starch digestibility (87%) and fecal excretion (21% of DM). Finer particles are quickly degradable in the rumen and might cause subacute acidosis reducing intake and performance (OWENS; SODERLUND, 2006). Rumen fermentation is less energy efficient (80%) due to heat losses and methane production. 7 The better energy efficiency of small intestine starch digestibility (97%) may compensate for the smaller rumen degradability rate of larger particles that go through it (HARMON; MCLEOD, 2001; HUNTINGTON; HARMON; RICHARDS, 2006). The small intestine (SI) in ruminants has limitations in amylase production and glucose absorption that contribute to the passage of intact starch and glucose towards the large intestine which is less energy efficient and increases starch rate in feces meaning in loss of potential energy for the ruminant (OWENS; SODERLUND, 2006). Therefore, excess of coarsely particles might be inefficient. Dry processing has shown a less efficiency in corn digestibility as values around 13 and 7% of starch remain indigestible even on how fine the grain is ground (GOUVÊA et al., 2019; SCHWANDT et al., 2016; ZINN; SHEN, 1998). HUNTINGTON; HARMON; RICHARDS, (2006) summarized that when more than 75% of starch in the small intestine is digestible there is an increase in the yield of digestible energy by shifting starch degradation in the rumen to be digestible in SI but values above 70% digestion were found only when 700g or less of starch was going to SI. 2.2.2. Ensiled corn Ensiling corn grain has a low storage cost and can be done in two ways, high moisture corn silage (HMC) and reconstituted corn grain silage (RCGS), both with satisfactory performance results in comparison with dry-processed corn (DA SILVA, 2016). Both methods lead to an increase in ruminal and total digestion, where flint corn may have the same digestibility values as ensiled dent corn. Increases in degradability are noticed to continue at least until 300 days of ensiling but it is more accentuated until 60 days (DA SILVA et al., 2019; FERNANDES, 2014; GERVÁSIO et al., 2023). The ruminal starch degradability was summarized as 86.55% for HMC and 99.25% for total tract starch digestion, avoiding almost 100% of losses (OWENS; SODERLUND, 2006). High moisture corn grain silage is made when corn is harvested between 60% and 65% of grains DM. Corn kernels are passed to a roll or grinder to damage the pericarp and expose the endosperm to enzymatic and bacterial attack in the silo. The advantage of this method is the no need for water addition before ensiling and lower 8 field losses due to early harvest, but, in contrast, there is a short harvest window to hit the grain's ideal moisture content. On a large scale, this short period can be a hindrance when an extensive amount of land must be harvested, also, moist grains force more the machine during harvest. Small farms might face problems of land availability to plant corn, which directs these farmers to the market. Farmers that buy corn to make HMC often report a large variety of moisture content and may need to add water in the silo. When a huge silo is used, the amount of grain available per day to unload on the farm may be an issue and might take too much time to close the silo, leading to aerobic deterioration and loss of water through evaporation. When dry corn is ground with water addition and then ensiled, it is called reconstituted corn grain silage or rehydrated corn grain silage. The advantage over the HMC is the strategic grain purchase in periods when it is cheaper, less DM variability regardless of resources, and no need for planting. In contrast with these advantages, a large amount of water is necessary to moisten the silo mass until it reaches 65~60% DM, which can be a limiting factor in regions with low amounts of water availability. JACOVACI et al., (2021) did a metanalysis evaluating the effects of ensiling flint corn grain on feedlot beef cattle and found positive results for DM intake (8.85 vs 10.3; kg), feed efficiency (0.194 vs 0.164), total tract starch digestibility (0.991 vs 0.959) and diet metabolizable energy (ME; 12.9 vs 11.7; MJ/kg DM) with no difference in weight gain (1.58 vs 1.61 kg/day). The similar gain with lower intake can be supported by the higher grain digestibility and higher ME that might activate chemical mechanisms of satiety earlier in comparison to dry corn. The higher rumen fermentation also increases the risk of acidosis in non-well-balanced diets, but, on the other hand, the inclusion of corn grain silage stimulates a decline in meal size and increases in number of meals. In conclusion, ensiled corn grain has a feed value improvement of 25.7% in comparison with dry corn grain (JACOVACI, 2019). Low rumen pH is found in animals eating diets with corn grain silage (GODOI et al., 2021). The risk of subacute acidosis can reduce intake and negatively affect performance (MILLEN et al., 2016). Formulating diets with different resources of starch may lead to a better synergy with different protein resources in the rumen and may reduce the risk of acidosis. STOCK; ERICKSON, (2006) summarized data from different experiments using ensiled corn and dry corn and found that the inclusion of 9 70% of ensiled corn could increase performance and efficiency. In contrast, COULSON et al., (2023) and GERVÁSIO, (2021) evaluated diets with 50:50 inclusion of HMC and dry corn and observed better efficiency for animals eating 100% roll HMC. For ground HMC, 100% inclusion of HMC provides similar efficiency to cattle eating 50:50 inclusion of rolled HMC and DRC (COULSON et al., 2023). In only one found study using RCGS for animals finished on pasture, Nellore bulls receiving ad libitum supplements containing RCGS had better performance and efficiency than the ones consuming dry- ground corn supplement (GONÇALVES, 2018). 2.2.2.1. Reconstituted corn grain silage To make reconstituted RCGS, corn must be ground or rolled, moistened, mixed and then compacted to achieve ideal conditions (e.g., moisture and anaerobic environment) so enzymes and microorganisms start consuming residual O2. In anaerobic conditions, bacteria added in the silo or even epiphytic from the corn kernel begin a fermentation process that reduces the pH and has a great effect on the breakdown of prolamins (JUNGES et al., 2017)(MCDONALD; HENDERSON; HERON, 1991). Bacteria are responsible for 60.4% of proteolysis in the corn grain silo, while 29.5% are attributed to kernel enzymes, and only 4.8% is resulted from acid activity (JUNGES et al., 2017). In addition to being responsible for proteolysis in the silo, bacteria are responsible for the fermentation and indirect responsible for aerobic stability of the silo post-opening. To ferment properly, reconstituted corn grain silos must be in ideal conditions for bacteria. Besides anaerobic conditions and grain processing, moisture content directly impacts silage quality. The best moisture content for ensiling grain to achieve a satisfactory fermentative profile, aerobic stability, and DM degradability is between 35 and 40% (GOMES et al., 2020). The grain silage particle size also affects the DM degradability and the retention of water as water may percolate and not be absorbed by the coarse particles, resulting in inferior silage moisture content. Grains particle size is in function of mill settings. Finer particles have larger contact surfaces and expose more starch granules to microbial and enzymatic attacks 10 but require more energy and time to be processed. Even though RCGS ferments slower and less extensively when compared to typical whole plant forage silage (TAYLOR; KUNG, 2002), it has an adequate fermentation profile independent of particle size (DA SILVA, 2015; GERVÁSIO, 2021) but longer storage periods are recommended for coarse silages, principally to have similar DM degradability results as fine ground silage (DA SILVA et al., 2019; GERVÁSIO, 2021). Storage time affects positively the aerobic stability and starch degradability. Proteolysis is noted to be constant during the storage time, but it is more intensive at the beginning of the period (DA SILVA et al., 2019; HOFFMAN et al., 2011). Greater gains in DM degradability in inoculated RCGS are found until 57 days of ensiling, even though it continues as the silo remains stable (DA SILVA et al., 2019). Longer storage times lead to the accumulation of fermentation products that kill undesirable microorganisms. Expressive increases in inoculated RCGS aerobic stability vary and are found between 120 and 240 days of storage, proximally (DA SILVA et al., 2019; GERVÁSIO et al., 2023). After this time, the substrate for acid production starts to decrease followed by the beneficial microorganism population and the increment in AS becomes less expressive. Longer storage periods can also be a problem for farmers who have a high rate of silage use and for those in times of feed deficit. The use of inoculants can be a strategy to accelerate and improve the desirable effects in corn grain ensiling. 2.3. Corn grain silage inoculants The use of inoculants in silage is strategic to correct some mistakes in the confection (e.g., moisture, compacting) but in the case of RCGS, it works as an insurance as it boosts fermentation and aerobic stability. Cereal silo is a huge amount of immobile capital and nutrients for yeast and molds, thus, additives become an indispensable tool. Inoculants are mainly composed of lactic acid bacteria, which are divided in three major groups according to carbohydrate metabolism: homofermentative bacteria, facultative heterofermentative, and obligatory heterofermentative bacteria. Homofermentative bacteria metabolize hexoses by glycolysis pathway producing 2 11 lactic acids as the only product of their metabolism. Facultative heterofermentative ferments hexoses like homofermentative bacteria and pentoses by phosphoketolase pathway which produces mainly acetic and lactic acid and ethanol. The third group uses hexoses and pentoses throw the phosphoketolase pathway (REIS et al., 2018). MORAIS et al., (2017) made a review comparing the characteristics of corn grain silage (CGS) inoculated with different inoculants: Corn grain silage is characterized as moderate fermentation reaching pH 4.42 when no inoculant is used, the addition of homolactic bacteria improves pH drop to 4.16 but reduces aerobic stability by 40 hours. Comparing non-inoculated CGS against silage treated with heterofermentative bacteria, the inoculated ones were more than 100 hours more stable and pH was equal but tended to increase the fermentative DM loss by 0.5 points; the combination of both types of bacteria increased the aerobic stability by 40 hours and decrease the pH by 0.15. The exclusive use of homolactic bacteria will direct sugars to be converted into lactic acid (pKa = 3.86) making a faster and deeper pH drop which avoids undesirable fermentation and decreases DM loss during fermentation. However, this acid can be used by yeast in the opening phase in which will increase pH, making the environment adequate for molds, leading to the loss of the aerobic stability. On the other hand, the exclusive use of obligatory heterofermentative bacteria produces less lactic acid slowing down the pH drop which gives time to secondary fermentation increasing DM losses, but, it produces weak acids (e.g. acetic and propionic) that will increase silage aerobic stability as they have antifungal activity (MOON, 1983). Weak acids, when are in their non-dissociate form, penetrate yeast, and mold membranes and dissociate in there leading to the death of the microorganisms (MCDONALD; HENDERSON; HERON, 1991). Tropical countries have higher temperatures, therefore, the use of obligatory heterofermentative bacteria makes more sense since undesirable microorganisms grow faster in higher temperatures, making silage spoiling easier. Meanwhile, a combination of the previously mentioned group plus facultative heterofermentative may be ideal in case of mistakes during ensiling (e.g., wrong moisture). Among facultative heterofermentative bacteria, we can highlight L. plantarum, however, the use of this bacteria alone in RCGS has not been showing positive results 12 Figure 4: A Glicolyse pathway. B Phosphketalse pathway. Adapted from Chung et al., (2021) in aerobic stability (da Silva et al., 2018; Saylor et al., 2020). In the obligatory heterofermentative group, L. buchneri is one of the most famous bacteria and the mainly used to preserve grain silage under aerobic exposure. Corn grain silages inoculated with L. buchneri start to show significant improvement in AS between 45 and 60 days of ensiling (KLEINSCHMIT; KUNG, 2006a; SCHMIDT et al., 2009), with even better effects after 60 days tending to stabilize around 150 days reaching around 250 h of AS (DA SILVA et al., 2019). In anaerobic conditions, L. buchneri has the capacity to convert lactic acid into acetic acid (OUDE ELFERINK et al., 2001). Another bacteria that belongs to the same group, L. hilgardii was first studied as an inoculant by ÁVILA et al., (2014) in sugar cane silage, being the most promising among 58 Pentose 2 13 strains of 3 different species. SANTOS et al., (2017) studying the same bacteria reported a similar fermentative profile to L. buchneri. In only one study found until the moment using L. hilgardii in corn grain silage, its inoculation showed good results in aerobic stability after 10 ensiled days and when it was inoculated with L. buchneri, the results were even better (DA SILVA et al., 2021) being an excellent tool in an emergency, when the silo must be opened before the minimum recommended time of 60 days and still with good results in AS. Few experiments evaluated the aerobic stability of cattle diets, but none of them included grain silage. BERNARDES et al., (2007) in laboratory conditions observed that combination of L. plantarum + Propianobacterium unsured aerobic stability of a diet made with Brachiaria silage for 6 days. BERNARDES et al., (2007) and TAYLOR et al., (2002) did not observe loss of aerobic stability for at least 120 hours in diets made with Brachiaria silage (pre wield to 30% MS) and whole plant barley silage, respectively, both inoculated with L. buchneri. Taylor et al., (2002), in field conditions observed a reduction in aerobic stability to 79 h for inoculated silage ration against 46 h for control ration. SEPPÄLÄ et al., (2013) demonstrated in their experiment how high hygienic ingredients can affect the aerobic stability of the ration as they got a 50 greater AS for ration made with good-hygienic ingredients and it could have been more, but they stop measuring. 2.4. Beef cattle finishing phase on pasture with high supplementation Beef cattle high supplementation fattening on pasture consists of providing supplemental levels above 1.2% of body weight (BW), allowing for high stocking rate and performance, reaching the slaughter point in approximately 120 days. This technical management aligns with short-cycle livestock that has been recommended in Brazil for producing better meat quality and enabling greater profitability per area. In the finishing phase on pasture, the forage mass comprises the dietary roughage to provide ruminal health, and the supplement is provided in a feed trough eliminating the necessity for feedlot pens (SIQUEIRA; RESENDE; MORETTI, 2014). It is important to highlight that the minimum forage requirement must be 14 available for the animals throughout the whole phase (GALYEAN; HUBBERT, 2014) and the amount might influence results. Cattle coming from the growing phase with a low forage allowance (1.9 kg DM/kg BW) have better average daily gain and feed efficiency when it is finished with a high forage allowance (2.9 kg DM/kg BW), however, forage allowance in finishing phase does not affect cattle coming from a growing phase with high forage allowance (starting at 4.1 DM kg BW) (MOTA et al., 2020). Cattle finished on pasture can perform similarly to cattle in feedlots. (MORETTI, 2015). The supplement provided is similar to the concentrate used in Brazilian feedlots where 97.22% utilize corn as the main grain in the diet (SILVESTRE; MILLEN, 2021). The dry corn grain used in the supplement may be rehydrated and ensiled serving as a low cost stock. One experiment conducted in APTA Colina evaluated high-moisture corn grain silage in supplements for finishing animals on pasture, completely substituting dry corn grain. The inclusion of high-moisture corn reduced supplement intake by 8.1 % (650g / day), increased weight gain by 120 g/day, and improved feed efficiency by 19% (151 vs 180 g). Even though forage intake was not measured, forage mass and stocking rate were greater in pasture for animals consuming HMCS which can be an effect in the reduction of pasture consumption due to the same mechanism of feed regulation applied in the reduction of the supplement intake (GONÇALVES, 2018) Cattle on pasture with low supplementation have larger rumen size and TGI fill compared with animals eating high-energy diets. When we compare finishing cattle on pasture against finishing animals on feedlot, animals on pasture have smaller rumen and TGI fill due to the possibility of selecting better quality forage and the amount ingested while animals on feedlot must consume the mixed diet (MORETTI, 2015). Smaller rumen size and TGI fill tend to lead to a better carcass dressing. A disadvantage of animals on pasture is the necessity to spend more energy to collect forage increasing net energy for maintenance while feedlot animals walk short distances saving more energy for gain. In general , studies show good efficiency for cattle finished on pasture with high supplementation being a solution for Brazilian livestock. 15 3. Objective and Hypothesis The experiment hypothesis is: RCGS inoculation with L. buchneri + L. hilgardii or L. buchneri + L. plantarum is capable of ensuring the aerobic stability of the RCGS and the supplement containing this silage for at least 72 h, preserving nutrients and ensuring performance similar to animals treated with a supplement provided immediately after preparation. Replacing 70% of DRC with RCGS will improve cattle efficiency and performance. The aims of this study are: I) to evaluate the combination of L. buchneri + L. hilgardii and L. buchneri + L. plantarum in the ensiling of reconstituted corn grains and their effects on silage fermentation, nutritional value, and aerobic stability of the silage and the supplement made with these silage; II) to determine the impacts of providing fresh and exposed supplement on the intake and performance of pasture-finished Nellore bulls with high supplementation. 4. References ALLEN, M. S.; BRADFORD, B. J.; OBA, M. BOARD-INVITED REVIEW: The hepatic oxidation theory of the control of feed intake and its application to ruminants. Journal of Animal Science, v. 87, n. 10, p. 3317–3334, 1 out. 2009. AMARAL, R. C. et al. Cana-de-açúcar in natura ou ensilada com e sem aditivos químicos: estabilidade aeróbia dos volumosos e das rações. Revista Brasileira de Zootecnia , v. 38, n. 10, p. 1857–1864, 2009. ÁVILA, C. L. S. et al. The use of Lactobacillus species as starter cultures for enhancing the quality of sugar cane silage. Journal of Dairy Science, v. 97, n. 2, p. 940–951, fev. 2014. BERNARDES, T.; CASTRO, T. PSXII-12 Silages and roughage sources in the Brazilian beef feedlots. Journal of Animal Science, v. 97, n. Suppl. 3, p. 411- (Abstr.), 2019. BERNARDES, T. F. et al. Estabilidade aeróbia da ração total e de silagens de capim- marandu tratadas com aditivos químicos e bacterianos. Revista Brasileira de Zootecnia, v. 36, n. 4, p. 754–762, 2007. BORREANI, G. et al. Silage review: Factors affecting dry matter and quality losses in silages. Journal of Dairy Science, v. 101, n. 5, p. 3952–3979, 1 maio 2018. 16 CAETANO, M. et al. Effect of flint corn processing method and roughage level on finishing performance of Nellore-based cattle. Journal of animal science, v. 93, n. 8, p. 4023–4033, 6 ago. 2015. CAETANO, M. et al. Impact of flint corn processing method and dietary starch concentration on finishing performance of Nellore bulls. Animal Feed Science and Technology, v. 251, p. 166–175, 1 maio 2019. CARVALHO, P. DE A. Influência do genótipo e maturiadde na diversidade microbiológica em milho grão para silagem. Piracicaba: Escola Superior de Agricultura “Luiz de Queiroz”, 2014. CEPEA, C. DE E. A. EM E. A. Centro de Estudos Avançados em Economia Aplicada. CORONA, L. et al. Comparative Effects of Whole, Ground, Dry-Rolled, and Steam- Flaked Corn on Digestion and Growth Performance in Feedlot Cattle. Professional Animal Scientist, v. 21, n. 3, p. 200–206, 1 jun. 2005. CORREA, C. E. S. et al. Relationship Between Corn Vitreousness and Ruminal In Situ Starch Degradability. Journal of Dairy Science, v. 85, n. 11, p. 3008–3012, 1 nov. 2002. COULSON, C. Evaluation of Grain Type and Processing Method on Steer Performance, Carcass Characteristics, and Nutrient Digestion. Lincoln: University of Nebraska - Lincoln, 2021. COULSON, C. A. et al. Impact of different corn milling methods for high-moisture and dry corn on finishing cattle performance, carcass characteristics, and nutrient digestion. Journal of Animal Science, v. 101, p. 1–10, 3 jan. 2023. DA SILVA, E. B. et al. Effects of Lactobacillus hilgardii 4785 and Lactobacillus buchneri 40788 on the bacterial community, fermentation and aerobic stability of high-moisture corn silage. Journal of Applied Microbiology, v. 130, n. 5, p. 1481–1493, 1 maio 2021. DA SILVA, É. B. et al. The use of Lentilactobacillus buchneri PJB1 and Lactiplantibacillus plantarum MTD1 on the ensiling of whole-plant corn silage, snaplage, and high-moisture corn. Journal of dairy science, v. 107, n. 2, fev. 2024. DA SILVA, N. C. Características das silagens de grãos de milho influenciadas pela reidratação e pela inoculação com L. buchneri sobre o desempenho de bovinos de corte confinados. Jaboticabal: Universidade Estadual Paulista (Unesp), 2016. DA SILVA, N. C. et al. Fermentation and aerobic stability of rehydrated corn grain silage treated with different doses of Lactobacillus buchneri or a combination of Lactobacillus plantarum and Pediococcus acidilactici. Journal of Dairy Science, v. 101, n. 5, p. 4158–4167, 1 maio 2018. 17 DA SILVA, N. C. et al. Influence of storage length and inoculation with Lactobacillus buchneri on the fermentation, aerobic stability, and ruminal degradability of high- moisture corn and rehydrated corn grain silage. Animal Feed Science and Technology, v. 251, p. 124–133, 1 maio 2019. DE VRIES, M. F. W. Estimating forage intake and quality in grazing cattle: A reconsideration of the hand-plucking method. Journal of Range Management, v. 48, n. 4, p. 370–375, 1995. DROUIN, P.; TREMBLAY, J.; CHAUCHEYRAS-DURAND, F. Dynamic succession of microbiota during ensiling of whole plant corn following inoculation with lactobacillus buchneri and lactobacillus hilgardii alone or in combination. Microorganisms, v. 7, n. 12, 1 dez. 2019. FERNANDES, J. Influência de genótipo, maturidade e tempo de armazenamento na qualidade de silagens de grãos de milho com alta umidade. Piracicaba: Universidade de São Paulo, 25 jul. 2014. FERRARETTO, L. F.; CRUMP, P. M.; SHAVER, R. D. Effect of cereal grain type and corn grain harvesting and processing methods on intake, digestion, and milk production by dairy cows through a meta-analysis. Journal of Dairy Science, v. 96, n. 1, p. 533–550, jan. 2013. FREITAS, T. B. et al. Effect of feeding dry-rolled corn or whole shelled corn during the finishing phase on growth performance and carcass characteristics. Translational Animal Science, v. 5, n. 1, p. 1–8, 1 jan. 2021. GALYEAN, M. L.; HUBBERT, M. E. Review: Traditional and alternative sources of fiber-Roughage values, effectiveness, and levels in starting and finishing diets. Professional Animal ScientistElsevier Inc., , 1 dez. 2014. GARCÍA-LARA, S.; CHUCK-HERNANDEZ, C.; SERNA-SALDIVAR, S. O. Development and Structure of the Corn Kernel. Em: Corn: Chemistry and Technology, Third Edition. 3. ed. [s.l.] Elsevier, 2019. p. 147–163. GAYRAL, M. et al. Transition from vitreous to floury endosperm in maize (Zea mays L.) kernels is related to protein and starch gradients. Journal of Cereal Science, v. 68, p. 148–154, 1 mar. 2016. GERVÁSIO, J. R. S. REIDRATAÇÃO E ENSILAGEM DE GRÃOS DE MILHO COM DIFERENTES GRANULOMETRIAS E INCLUSÕES NA DIETA PARA BOVINOS DE CORTE. Jaboticabal: Universidade Estadual Paulista, 2021. GERVÁSIO, J. R. S. et al. Effects of particle size and storage length on the fermentation pattern and ruminal disappearance of rehydrated corn grain silage hammer mill processed. Animal Feed Science and Technology, v. 306, p. 115810, 1 dez. 2023. 18 GODOI, L. A. et al. Effect of flint corn processing methods on intake, digestion sites, rumen pH, and ruminal kinetics in finishing Nellore bulls. Animal Feed Science and Technology, v. 271, p. 114775, 1 jan. 2021. GOMES, A. L. M. et al. Effects of processing, moisture, and storage length on the fermentation profile, particle size, and ruminal disappearance of reconstituted corn grain. Journal of Animal Science, v. 98, n. 11, 2020. GONÇALVES, P. H. AVALIAÇÃO DO PROCESSAMENTO DO MILHO E DA LASALOCIDA NA TERMINAÇÃO DE BOVINOS NELORE EM SISTEMA DE ALTA SUPLEMENTAÇÃO NA SECA. Jaboticabal: Universidade Estadual Paulista “Júlio de Mesquita Filho” - Campus Jaboticabal, 2018. GOROCICA-BUENFIL, M. A.; LOERCH, S. C. Effect of cattle age, forage level, and corn processing on diet digestibility and feedlot performance. Journal of Animal Science, v. 83, n. 3, p. 705–714, 1 mar. 2005. GOUVÊA, V. N. et al. Effects of alternative feed additives and flint maize grain particle size on growth performance, carcass traits and nutrient digestibility of finishing beef cattle. The Journal of Agricultural Science, v. 157, n. 5, p. 456–468, 1 jul. 2019. HARMON, D. L.; MCLEOD, K. R. Glucose uptake and regulation by intestinal tissues: Implications and whole-body energetics. Journal of Animal Science, v. 79, n. suppl_E, p. E59–E72, 1 jan. 2001. HOFFMAN, P. C. et al. Influence of ensiling time and inoculation on alteration of the starch-protein matrix in high-moisture corn. Journal of Dairy Science, v. 94, n. 5, p. 2465–2474, maio 2011. HOFFMAN, P. C.; OCHER, S. M. Quantification of milk yield losses associated with aerobically unstable high moisture corn. Journal of Dairy Science, v. 80, n. Suppl. 1, p. 234- (Abstr.), 1997. HUNTINGTON, G. B.; HARMON, D. L.; RICHARDS, C. J. Sites, rates, and limits of starch digestion and glucose metabolism in growing cattle. Journal of Animal Science, v. 84, n. suppl_13, p. E14–E24, 1 abr. 2006. JACOVACI, F. A. et al. Effect of ensiling on the feeding value of flint corn grain for feedlot beef cattle: A meta-analysis. Revista Brasileira de Zootecnia, v. 50, n. e20200111, 25 fev. 2021. JOBIM, C. C. et al. Avanços metodológicos na avaliação da qualidadeda forragem conservada. Revista Brasileira de Zootecnia, v. 36, p. 101–119, 2007. JUNGES, D. et al. Short communication: Influence of various proteolytic sources during fermentation of reconstituted corn grain silages. Journal of Dairy Science, v. 100, n. 11, p. 9048–9051, 1 nov. 2017. 19 KLEINSCHMIT, D. H.; KUNG, L. The effects of Lactobacillus buchneri 40788 and Pediococcus pentosaceus R1094 on the fermentation of corn silage. Journal of Dairy Science, v. 89, n. 10, p. 3999–4004, 2006a. KLEINSCHMIT, D. H.; KUNG, L. A meta-analysis of the effects of Lactobacillus buchneri on the fermentation and aerobic stability of corn and grass and small-grain silages. Journal of Dairy Science, v. 89, n. 10, p. 4005–4013, 1 out. 2006b. KLEINSCHMIT, D. H.; KUNG, L. The effects of Lactobacillus buchneri 40788 and Pediococcus pentosaceus R1094 on the fermentation of corn silage. Journal of Dairy Science, v. 89, n. 10, p. 3999–4004, 1 out. 2006c. KUNG, L. et al. The effect of Lactobacillus buchneri 40788 on the fermentation and aerobic stability of ground and whole high-moisture corn. Journal of Dairy Science, v. 90, n. 5, p. 2309–2314, 2007. KUNG Jr, L. Aerobic stability of silage. In: Proceedings, 2010 California Alfalfa & Forage Symposium and Corn/Cereal Silage Conference, Visalia, CA, 1-2 December, 2010. MCDONALD, P.; HENDERSON, A. R.; HERON, S. J. E. The biochemistry of silage. Em: WEBSTER, J. (Ed.). Experimental Agriculture. 2. ed. [s.l: s.n.]. v. 28p. 340–340. MILLEN, D. D. et al. Ruminal Acidosis. Em: Ruminlogy. [s.l: s.n.]. p. 127–156. MOON, N. J. Inhibition of the growth of acid tolerant yeasts by acetate, lactate and propionate and their synergistic mixtures. Journal of Applied Bacteriology, v. 55, p. 453–460, 1983. MORAIS, G. et al. Additives for grain silages: A review. Slovak Journal of Animal Science, v. 50, n. 1, p. 42–54, 2017. MORETTI, M. H. ESTRATÉGIAS ALIMENTARES PARA A RECRIA E TERMINAÇÃO DE TOURINHOS NELORE. Jaboticabal: Unesp - FCAV, 2015. MOTA, V. A. C. et al. Relationship between gain rate during the growing phase and forage allowance in the finishing phase in Nellore cattle. Tropical Animal Health and Production, v. 52, n. 4, p. 1881–1891, 1 jul. 2020. MUCK, R. E. et al. Silage review: Recent advances and future uses of silage additives. Journal of Dairy Science, v. 101, n. 5, p. 3980–4000, 1 maio 2018. NATIONAL ACADEMIES OF SCIENCES ENGINEERING AND MEDICINE. Nutrient Requirements of Beef Cattle: Eighth Revised Edition. Washington, DC: The National Academies Press, 2016. NIKKHAH, A. Barley grain for ruminants: A global treasure or tragedy. Journal of animal science and biotechnology, v. 3, n. 1, 9 jul. 2012. 20 NUÑEZ, A. J. C. et al. Combined use of ionophore and virginiamycin for finishing Nellore steers fed high concentrate diets. Scientia Agricola, v. 70, n. 4, p. 229–236, jul. 2013. OUDE ELFERINK, S. J. W. H. et al. Anaerobic conversion of lactic acid to acetic acid and 1,2-propanediol by Lactobacillus buchneri. Applied and Environmental Microbiology, v. 67, n. 1, p. 125–132, 2001. OWENS, F. N. et al. The effect of grain source and grain processing on performance of feedlot cattle: a review. Journal of animal science, v. 75, n. 3, p. 868–879, 1997. OWENS, F. N.; BASALAN, M. Grain processing: gain and efficiency responses by feedlot cattle. Proceedings of the Plains Nutrition Council Spring Conference. Anais...Amarillo, TX: 2013. OWENS, F.; SODERLUND, S. RUMINAL AND POSTRUMINAL STARCH DIGESTION BY CATTLE. Cattle Grain Processing Symposium. Anais...Tusla: 2006. PEREIRA, M. N. et al. Ruminal degradability of hard or soft texture corn grain at three maturity stages. Scientia Agricola, v. 61, n. 4, p. 358–363, 1 jul. 2004. PEREIRA, R. C. et al. Relationship between structural and biochemical characteristics and texture of corn grains. Genetics and molecular research : GMR, v. 7, n. 2, p. 498–508, 2008. PHILIPPEAU, C.; LANDRY, J.; MICHALET-DOREAU, B. Influence of the protein distribution of maize endosperm on ruminal starch degradability. Journal of the Science of Food and Agriculture, v. 80, p. 404–408, 2000. REIS, C. B. et al. Wild Lactobacillus hilgardii (CCMA 0170) strain modifies the fermentation profile and aerobic stability of corn silage. Journal of Applied Animal Research, v. 46, n. 1, p. 632–638, 1 jan. 2018. REIS, R. A. et al. Semi-confinamento para produção intensiva de bovinos de corte. I Simpósio Matogrossense de bovinocultura de corte. Anais...2011. ROONEY, L. W.; PFLUGFELDER, R. L. Factors affecting starch digestibility with special emphasis on sorghum and corn. Journal of animal science, v. 63, n. 5, p. 1607–1623, 1986. SANTOS, W. P. et al. Effect of the inoculation of sugarcane silage with Lactobacillus hilgardii and Lactobacillus buchneri on feeding behavior and milk yield of dairy cows. Journal of Animal Science, v. 95, n. 10, p. 4613–4622, 1 out. 2017. SAYLOR, B. A. et al. Effect of microbial inoculation and particle size on fermentation profile, aerobic stability, and ruminal in situ starch degradation of high-moisture corn ensiled for a short period. Journal of Dairy Science, v. 103, n. 1, p. 379–395, 1 jan. 2020. 21 SCHMIDT, R. J. et al. The development of lactic acid bacteria and Lactobacillus buchneri and their effects on the fermentation of alfalfa silage. Journal of Dairy Science, v. 92, n. 10, p. 5005–5010, 2009. SCHWANDT, E. F. et al. The effects of dry-rolled corn particle size on performance, carcass traits, and starch digestibility in feedlot finishing diets containing wet distiller’s grains. Journal of animal science, v. 94, n. 3, p. 1194–1202, 1 mar. 2016. SEPPÄLÄ, A. et al. Controlling aerobic stability of grass silage-based total mixed rations. Animal Feed Science and Technology, v. 179, n. 1–4, p. 54–60, 31 jan. 2013. SILVESTRE, A. M.; MILLEN, D. D. The 2019 brazilian survey on nutritional practices provided by feedlot cattle consulting nutritionists. Revista Brasileira de Zootecnia, v. 50, p. 1–25, 2021. SIQUEIRA, G. R.; RESENDE, F. D.; MORETTI, M. H. Terminação de bovinos inteiros em pastagens. Pesquisa & Tecnologia, v. 11, n. 1, 2014. STOCK, R. A.; ERICKSON, G. E. Associative Effects and Managements – Combinations of Processed Grains. Cattle Grain Processing Symposium. Anais...Tulsa, Oklahoma: 2006. TAYLOR, C. C. et al. The effect of treating whole-plant barley with Lactobacillus buchneri 40788 on silage fermentation, aerobic stability, and nutritive value for dairy cows. Journal of Dairy Science, v. 85, n. 7, p. 1793–1800, 2002. TAYLOR, C. C.; KUNG, L. The effect of Lactobacillus buchneri 40788 on the fermentation and aerobic stability of high moisture corn in laboratory silos. Journal of Dairy Science, v. 85, n. 6, p. 1526–1532, 2002. TRICARICO, J. M. et al. Effects of a dietary Aspergillus oryzae extract containing α- amylase activity on performance and carcass characteristics of finishing beef cattle. Journal of Animal Science, v. 85, n. 3, p. 802–811, 1 mar. 2007. VANDER POL, K. J. et al. Effect of Corn Processing in Finishing Diets Containing Wet Distillers Grains on Feedlot Performance and Carcass Characteristics of Finishing Steers1. Professional Animal Scientist, v. 24, n. 5, p. 439–444, 1 out. 2008. WEISS, W. P.; CONRAD, H. R.; ST. PIERRE, N. R. A theoretically-based model for predicting total digestible nutrient values of forages and concentrates. Animal Feed Science and Technology, v. 39, n. 1–2, p. 95–110, 16 nov. 1992. WHEELER, W. E.; NOLLER, C. H. Gastrointestinal Tract pH and Starch in Feces of Ruminants. Journal of Animal Science, v. 44, n. 1, p. 131–135, 1 jan. 1977. WHITLOCK, L. A. et al. Effect of level of surface-spoiled silage on the nutritive value of corn silage-based rations. Kansas Agricultural Experiment Station Research Reports, n. 1, p. 22–24, 1 jan. 2000. 22 WILCOX, R. A.; DEYOE, C. W.; PFOST, H. B. A Method for Determining and Expressing the Size of Feed Particles by Sieving. Poultry Science, v. 49, n. 1, p. 9– 13, 1 jan. 1970. ZHANG, H.; GAO, R.; DONG, S. Anatomical and Physiological Characteristics Associated with Corn Endosperm Texture. Agronomy Journal, v. 103, n. 4, p. 1258– 1264, 1 jul. 2011. ZHANG, H.; XU, G. Physicochemical properties of vitreous and floury endosperm flours in maize. Food Science and Nutrition, v. 7, n. 8, p. 2605–2612, 2019. ZINN, R. A. et al. Starch digestion by feedlot cattle: Predictions from analysis of feed and fecal starch and nitrogen. Journal of Animal Science, v. 85, n. 7, p. 1727–1730, jul. 2007. ZINN, R. A.; OWENS, F. N.; WARE, R. A. Flaking corn: processing mechanics, quality standards, and impacts on energy availability and performance of feedlot cattle. Journal of animal science, v. 80, n. 5, p. 1145–1156, 2002. ZINN, R. A.; SHEN, Y. An Evaluation of Ruminally Degradable Intake Protein and Metabolizable Amino Acid Requirements of Feedlot Calves. Journal of Animal Science, v. 76, p. 1280–1289, 1998 23 CHAPTER 2 – EXPERIMENT 1: EVALUATION OF SILAGE INOCULANTS FOR RECONSTITUTED CORN GRAIN SILAGE AND SUPPLEMENT AEROBIC STABILITY. 1. Introduction In 2019, 32% of feedlots in Brazil used any type of matured grain silage (e.g HMC, RCGS, snaplage) (BERNARDES; CASTRO, 2019) and this number might increase as the benefits of these technologies become more known among producers. Hence, producers using other strategies to fatten cattle might want to use it too. It is recommended to unload the silo daily and to use the silage immediately after unloading as it can start to spoil quicker but in some business models, some of these recommendations may not be followed and extra care must be taken. Dry corn kernels used to make RCGS have a low concentration of water- soluble carbohydrates, a low count of desirable epiphytic LAB, and may have high count of undesirable microorganisms (CARVALHO, 2014). This combination can be a factor that will imply poor fermentation, high losses, and short aerobic stability. Aerobic spoilage is a phenomenon where the silage mass has contact with air and aerobic microorganisms start to multiply and consume nutrients (Pahlow et al., 2003). This is one of the main aggravating factors in silage production. With the advance of studies on silage fermentation and conservation, technologies were developed to control these problems, and DM loss due to poor fermentation and spoilage after opening becomes inadmissible, principally in cases where the silage will be exposed for some time for sure. Facultative heterofermentative bacteria are indicated for silage with fermentation problems because they produce mainly lactic acid and obligatory heterofermentative bacteria are indicated to improve aerobic stability as weak acids are the main product of their fermentation. Even with low substrate to be fermented and a smaller capacity to decrease pH obligatory heterofermentative bacteria are capable of providing a good fermentation profile in RCGS silage (DA SILVA et al., 2021, 2024, 2018, 2019). L. Buchneri is one of the most studied bacteria in this 24 group of obligatory heterofermentative bacteria and shows excellent improvement in aerobic stability (KLEINSCHMIT; KUNG, 2006c; KUNG et al., 2007; TAYLOR; KUNG, 2002). L. Hilgardii is in the same group and it seems to have a similar metabolism that L. Buchneri with an earlier start of actuation (ÁVILA et al., 2014; DA SILVA et al., 2021; SANTOS et al., 2017). A combination of facultative and obligatory heterofermentative bacteria can be an option to guarantee fermentation and aerobic stability, and L. Plantarum is the main bacteria studied with good results to represent the facultative heterofermentative group(DA SILVA et al., 2024). When silage cannot be immediately used, it can be mixed with other ingredients and the ration or supplement will have to remain stable until being provided. In this situation, acids in the silage will reduce concentration, and undesirable microorganisms will increase due to the addition of other ingredients and contact with machines (SEPPÄLÄ et al., 2013). On the other hand, DM will increase, and water activity decrease, making the environment less favorable to microbial growth (JOBIM et al., 2007). All of these factors affect the aerobic stability of the supplement or ration, and studies are requested to understand this technique and see how it will work. This study aims to evaluate the combination of L. buchneri + L. hilgardii and L. buchneri + L. plantarum in the ensiling of reconstituted corn grains and their effects on silage fermentation, nutritional value, and aerobic stability of the silage and the supplement made with those silage. The hypothesis is inoculation with those bacteria will guarantee aerobic stability of the supplement for at least 72 hours (3 days). 2. Material and Methods The Experiment was conducted at the São Paulo State Agency for Agribusiness Technology – Alta Mogiana Polo, Colina – SP from October 2022 to December 2022. It was evaluated RCGS inoculated with the following treatments: RCGS without inoculation (CON); RCGS inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH, Lallemand Inc); RCGS inoculated 25 with L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB, Volac International Lim.). At the silo opening, a supplement was prepared using the silage and it was evaluated. 2.1. Ensiling Process and Supplement Preparation Dry corn grain was acquired from the local market, ground in a hammer mill (TN 9) with a 5 mm sieve and stored in a mixer wagon. A sample was collected to check the dry matter (DM) in a forced ventilation oven at 105° C for 12 h. With the dry matter result, approximately 115 kg of ground corn were transferred to a concrete mixer and chlorine-free water at room temperature was added and mixed for 5 minutes to increase mass moisture to 37.5%. Six piles of 25 kg of rehydrated corn were made above a pre-cleaned canvas. The inoculant was diluted in the same water used to rehydrate the corn and each pile was inoculated using a hand spray filled with 300 ml of diluted inoculant (12 l / ton). The pre-ensiled mass was hand-homogenized while it was inoculated. Control treatment received only water. After homogenization, the rehydrated corn from each treatment was sampled and compacted in pre-cleaned and weighed 23 L plastic buckets to achieve 1050 kg / m³, sealed with a lid and plastic tape and then weighed. This process was repeated 3 times and rehydrated corn from each repetition was sampled for analysis. The experimental silos were stored for 60 days. After the storage period, the buckets were weighed again and opened. Approximately 2 cm of silage from the top were discarded and the remaining was homogenized and sampled for further analysis. Three kilograms of silage were taken for aerobic stability evaluation. The remaining content from each bucket was individually used in the preparation of a supplement similar to that would be provided to cattle in experiment 2. The supplement consists of 20.19% ground dry corn, 47.12% RCGS, 11% cottonseed, 17.4% cottonseed cake, and 4.29% mineral premix (Table 2, Exp. 2). The supplement was manually mixed on a tarp and sampled for analysis. After mixed, 10 kg of the supplement were allocated to 2 new buckets (5 kg/bucket) previously identified, cleaned, and weighed with a data logger (Pro 2.07.09, Escort 26 Console, Buchanan, VA, United States) inside the bucket, totaling 36 buckets (2 supplement buckets/silage bucket). 2.2. Fermentation losses, aerobic stability, and aerobic exposure losses After 60 days of ensiling, the silos were weighed again to determine dry matter losses during the fermentation process (Jobim et al., 2007). Approximately 3 kg of silage was placed in plastic bags covered with sheet paper to prevent external contamination (Tabacco et al., 2009). The trays were kept in a closed room at a controlled temperature of 25° C for 9 days. The silage temperature was recorded every 30 minutes using a data logger positioned in the geometric center of the silage. Aerobic stability was defined as the number of hours the silage remained stable before reaching 2°C above room temperature (Ranjit & Kung, 2000). The buckets with supplements were kept open in the same temperature- controlled room and it was checked if aerobic stability was maintained for 72 h (3 days) or 144 h (6 days), following the same protocol of Ranjit & Kung, (2000) described in the previous paragraph. The buckets were weighed after the predetermined exposure time for calculations of dry matter losses due to aerobic exposure (Jobim et al., 2007). 2.3. Chemical analyses A pre-ensiled corn sample from each treatment was dried in a forced ventilation oven at 55°C for 72 h to estimate partial dry matter (DM) and then ground in a Willey mill using a 1 mm mesh sieve. The ground samples were used for quantification of dry matter in a 105°C oven for 12 h (method 934.01, AOAC, 2012) and total nitrogen by the Kjeldahl method (method 981.10, AOAC, 2012), crude protein (CP) content was calculated by multiplying total nitrogen content by 6.25. Water-soluble carbohydrates (Hall et al., 2015) was also quantified. Another pre-ensiled sample from each treatment was added to plastic bags containing distilled water at a 1:10 ratio and then homogenized in a Stomacher® device (400 Circulator, West Sussex, UK) for 4 minutes. From the solution, pH was checked using a potentiometer (DM-22, Digimed, 27 São Paulo, SP, Brazil) and ammonia nitrogen (N-NH3) content by following the procedures described in the INCT (Detmann et al., 2012). Silage sample was collected in each bucket on the silo opening day and dried in a forced ventilation oven at 55°C for 72 h to estimate partial DM and then ground in a Willey mill with a 1 mm mesh sieve. The ground samples were used for DM quantification, CP, and sugars all following the protocols described above. Additionally, ash analysis (method 930.05, AOAC, 2012), fats (method 945.16, AOAC, 2006) and starch (Hall et al., 2015) were performed. Another sample collected from each bucket was diluted with distilled water (1:10) and homogenized in a stomacher® for 4 minutes, then pH and N-NH3 were analyzed following the previously described protocols; additionally, the levels of alcohols, esters, and volatile fatty acids were analyzed using gas chromatography with mass detector (GCMS) (GCMS QP 2010 plus, Shimadzu®, Kyoto, Japan), with a capillary column (Stabilwax, Restek®, Bellefonte, USA; 60 m, 0.25 mm ø, 0.25 μm crossbond carbowax polyethylene glycol) and analytical parameters as recommended by the manufacturer. Samples of the supplement collected at the time of preparation and from each bucket after the aerobic exposure period were analyzed for DM, CP, fat, ash, sugars, starch, and pH. All analyses will follow the protocols referenced at the beginning of this section. Analyses of neutral detergent fiber with heat-stable amylase (NDF) and acid detergent fiber (ADF) were performed in the supplement samples (Van Soest et al., 1991) and total digestible nutrients (TDN) (NASEM, 2001). Table 1: Silo density and chemical composition of pre-ensiled reconstituted corn without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) Itema CON LBLH LPLB Silo Density (ton/m³) 1.04 1.03 1.03 DM (g/kg DM) 619 619 620 CP (g/kg DM) 98.4 98.0 98.2 WSC (g/kg DM) 16.1 16.1 16.1 pH 5.99 6.00 6.03 N-NH3(g/kg DM) 0.19 0.15 0.15 aDM = dry mater; CP = crude protein; WSA = water soluble carbohydrates; N-NH3 = ammonia nitrogen 28 2.4. Microbiological Analyses Samples collected from pre-ensiled material, post-opening silage, supplement ingredients, and supplement at times 0, 72, and 144 were placed in sterile bags and diluted with peptone water (1 g/L, 1:10 peptone water) and then homogenized for 4 minutes in a stomacher®. Surface plating technique on YGC Agar medium was carried out for yeast and filamentous fungi counting. Serial dilutions (10-1 to 10-6) were prepared and plated in duplicate (Tabacco et al., 2009). After incubation at 28° C for three and five days for yeast and filamentous fungi, respectively, colonies were counted separately, based on their macro-morphological characteristics. The same technique described for yeast and filamentous fungi was used for LAB counting, with MRS Agar medium and incubation at 35°C for three days for subsequent counting. Table 2: Microbial counts of supplement ingredients and pre-ensiled corn without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) Itema CON LBLH LPLB DGC Cottonseed cake Cottonseed whole LAB (log CFU/g) 3.71 4.11 4.46 - - - Yeast (log CFU/g) 2.82 3.05 3.01 3.11 2.59 2.30 Molds (log CFU/g) 2.91 3.10 3.23 4.08 3.18 2.44 aLAB: Lactic acid bacteria; CFU, colony-forming unity DGC: dry ground corn 2.5. In vitro and In vivo degradability and microscopy analyses Samples of RCGS, and supplement at all 3 times of evaluation were collected and dried in forced-air ovens at 55°C for 72 hours and then ground to 2 mm for in-vitro digestibility tests at a commercial laboratory (IFM® Alltech, Maringá – PR; Meads et al., 2021). The total gas production, and volatile fatty acids (VFA) production, were evaluated for 48 hours. Rumen fluid was collected from a Holstein cow consuming whole plant corn silage and minerals. Five grams of dry silage from each bucket, in the original particle size, were placed in pre-weighed nylon bags to analyze DM degradability for 12 hours in 29 canulated Nellore bulls receiving a diet of 15% Brachiaria hay, 68% rolled corn, 7% soybean meal, 7% citrus pulp, and 3.1% mineral supplement. After 12 hours, bags were taken out and rinsed with water until have gotten crystal clear water and then dried in an oven at 105°C for 12 hours. DM degradability was calculated as: (post- incubated dried total bag weight – bag weight) / [(pre-incubated total bag weight – bag weight) x sample DM %]. Furthermore, dry 1mm ground samples from pre-ensiled corn and post ensiled corn of each treatment were taken to The University of Adelaide and visually analyzed in a scanning electron microscopy model Ultim Max 65 (Oxford Instruments, Oxon - UK) to observe starch granules following recommendations described by Hoffman et al., (2011). 2.6. Statistical analysis Corn silage data were analyzed as a randomized block design using the MIXED procedure of SAS (SAS Institute, INC., Cary, NC). Each one of the 3 rehydrations of ground corn was considered as one block. The model considered treatment as fixed effect and block as random effect. Means were confronted in orthogonal contrasts for 5% probability: C1: CON vs LBLH + LPLB; C2: LBLH vs LPLB. For the supplement, data were analyzed as repeated measures using the MIXED procedure of SAS. Treatment, day, and treatment*day were considered fixed effects, block and bucket were considered random effects and bucket(day) was the subject. The following covariance structures: ANTE1 AR1 ARH1 ARMA11 CS CSH FA1 FA2 HF TOEP2 TOEPH2 UN UN1 UN2 VC were tested for all parameters. The covariance structure with the lowest Bayesian information criteria was chosen separately for each parameter. All means were compared using the t-test at a 5 % probability. Polynomial contrasts (linear and quadratic) were performed for days and C1 and C2 for treatments when no interaction effect was detected. When the interaction effect was significant, contrast for days (linear and quadratic) in treatments and treatments (C1 and C2) in days were performed. The significance level adopted was P < 0.05 and P ≤ 10 was considered tendency. 30 3. Results DM loss was two times higher for inoculated silages (C1, P < 0.01) as pH was 0.5 points higher for these silages (C1, P < 0.01). No differences for both characters were found in inoculated treatment silages (C2, P ≥ 0.11). Counting for LAB was approximately 8.08% higher in inoculated treatments (C1, P < 0.01) and similar for both inoculated treatments (C2, P < 0.01). Molds were not detected in all treatments and yeast was noted only in control silage (C1, P < 0.01). DM degradability at 12 hours was higher in inoculated silages (C1, P < 0.01) and higher in LBLH than in LBLP (C2, P < 0.01). N-NH3 was higher in inoculated silages (C1, P < 0.01) and tended to be higher in LBLH treatment (C2, P = 0.08) (Table 3). Table 3: DM loss, pH, microbial counts, dry matter digestibility and N-NH3 of RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) on the opening day. Itema Treatments SEM P-Valueb CON LBLH LBLP C1 C2 DM loss (g/kg DM) 22.4 47.8 42.4 0.13 <0.01 0.11 pH 3.95 4.45 4.45 0.01 <0.01 0.90 LAB (log CFU/g) 5.63 6.13 6.04 0.09 <0.01 0.52 Yeast (log CFU/g) 3.56 ND ND 0.07 <0.01 1.00 Molds (log CFU/g) ND ND ND 0.12 1.00 1.00 DMD 12h (g/kg DM) 745 782 757 0.35 <0.01 <0.01 N-NH3 (g/ kg DM) 0.91 1.40 1.35 0.19 <0.01 0.08 aDMD = dry matter degradability; LAB = lactic acid bacteria; DM loss = DM loss during storage period. bC1: CON vs LBLH + LPLB; C2: LBLH vs LPLB ND = counts lower than 2 CFU The production of lactic acid was 3.7 times greater for CON silage and differed from inoculated silages (C1, P < 0.01) which were equal between inoculated ones (C2, P = 0.13). Acetic acid were 4.15 times greater for inoculated silages (C1, P < 0.01) and LBLH produced 1.3g/kg DM more than LPLB (C2, P < 0.01). Butyric acid and ethanol had higher concentration in CON silage than in inoculated silages (C1, P < 0.01) and were similar in both inoculated silages (C2, P ≥ 0.14).1,2 propanediol was 9 times more 31 concentrate in inoculated silages than in non-inoculated silage (C1, P < 0.01) and LPLB had greater concentration than LBLH (C2, P < 0.01). Table 4: Fermentation products of RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) on the opening day. Iten (g/kg DM) Treatments SEM P-Valuea CON LBLH LPLB C1 C2 Lactic acid 27.6 7.03 7.88 0.71 <0.01 0.13 Acetic acid 2.95 12.9 11.6 0.17 <0.01 <0.01 Butyric acid 0.06 0.01 0.01 0.01 <0.01 0.41 Ethanol 2.95 2.33 2.18 0.07 <0.01 0.14 1,2 propanediol 0.348 2.67 3.76 0.28 <0.01 <0.01 aC1: CON vs LBLH + LPLB; C2: LBLH vs LPLB Aerobic stability was more than four times greater in inoculated treatments (C1, P < 0.01) and it was not lost in inoculated silages. DM after 9 days was greater in inoculated treatments (C1, P < 0.01) and similar in both inoculated (C2, P < 0.75). DM loss in this period was 6 times higher in non-inoculated silage (C1, P < 0.01) and similar in both inoculated (C2, P=0.70). Yeast and Molds were detected in only non-inoculated silage after 9 days (C1, P < 0.01) (Table 5). Table 5: Aerobic stability, DM, DM loss, microbial counts, and pH of RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) after AS period. Item Treatments SEM P-valueb CON LBLH LBLP C1 C2 AS (hours) 49.8 >216 >216 0.93 <0.01 1.00 DM (g/kg) 610 629 630 0.01 <0.01 0.75 DM loss (g/kg DM) 120 19.6 20.6 0.01 <0.01 0.70 pH 4.98 4.48 4.52 0.05 <0.01 0.62 Yeast (log CFU/g) 5.96 ND ND 0.06 <0.01 1.00 Molds (log CFU/g) 5.46 ND ND 0.07 <0.01 0.67 a AS = aerobic stability; DM = dry matter; DM loss = DM loss during exposing period. bC1: CON vs LBLH + LPLB; C2: LBLH vs LPLB ND: counts lower than 2 CFU 32 Figure 5: Temperature during aerobic stability test of RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB). DM at the opening was higher for non-inoculated silage (C1, P = 0.06) and similar for inoculated ones (C2, P = 0.64). PB was slightly higher in inoculated silages (C1, P < 0.01) and was higher in LPLB silage than in LBLH (C2, P < 0.01). Soluble protein was similar in inoculated and non-inoculated silage (C1, P = 0.52) and tended to be lower in LBLP compared to LBLH (C2, P = 0.07). Starch was similar for all silages (P ≥ 0.21) and sugar was lower in inoculated silages (C1, P < 0.01). Ash, fat, and TDN were similar for all treatments (P ≥ 0.12) (Table 6). Total gas production and VFA production did not differ between inoculated silages and control silage (P ≥ 0.16) (Table 7). Pre ensiled starch granules looks more aggregate than starch granules after storage period. In the 0d image is possible to notice the dense protein matrix unifying starch granules. After storage period, the protein matrix is reduced and fragmented, what facilitate microbial remen access and starch degradation (Figure 6) 22.00 27.00 32.00 37.00 42.00 47.00 0 20 40 60 80 100 120 140 160 180 200 220 T e m p e ra tu re ( ° C ) Hours Room temperature CON LBLH LPLB 33 Table 6: Chemical composition of RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) on the opening day. Itema Treatments SEM P-valueb CON LBLH LBLP C1 C2 DM (g/kg) 601 598 601 2.03 0.06 0.64 CP (g/kg DM) 100 103 109 1.51 <0.01 <0.01 SP (g/kg CP) 336 348 327 30.7 0.52 0.07 Starch (g/kg DM) 742 752 726 8.56 0.82 0.21 WSC (g/kg DM) 5.41 3.66 4.02 0.41 0.01 0.57 Fat (g/kg DM) 27.6 30.6 28.8 0.21 0.34 0.46 Ash (g/kg DM) 12.5 12.4 12.0 0.42 0.56 0.53 TDN (g/kg DM) 872 881 873 3.33 0.23 0.12 aDM = dry mater; CP = crude protein; SP = soluble protein; WSC = water soluble carbohydrates; TDN = total digestible nutrients. bC1: CON vs LBLH + LPLB; C2: LBLH vs LPLB Table 7: Total gas and VFA production of RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) on the opening day. Item Treatments P-valueb CON LBLH LBLP SEM C1 C2 Total gas production (ml/g DM) 209 212 221 6.72 0.16 0.17 Total VFAa (mM/g DM) 27.77 28.87 28.99 1.42 0.32 0.92 aVFA = volatile fat acids bC1: CON vs LBLH + LPLB; C2: LBLH vs LPLB 34 Figure 6: Starch granules of pre ensiled corn (0d), no inoculated RCGS (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) on the opening day. Control supplement remained stable when exposed to air for 107.5 hours while inoculated treatments remained stable for the whole experimental period (Figure 7). DM loss was similar between treatments on day 3 (P > 0.18) and differ for non- inoculated supplement in day 6 (P = 0.04) it also differs between day 3 and day 6 (P < 0.01) (Figure 8). pH, yeasts, and molds had an interaction between days and treatment (P < 0.01) (Figures 9, 10 and 11). pH differs between inoculated and non-inoculated treatments and type of inoculant on day 0 and between inoculated and non-inoculated supplement on day 6 (P < 0.01). In all days, all treatments had a linear and quadratic 0d CON LBLH LPLB 35 effect (P < 0.01). Yeasts were higher in all days for control (P> 0.01) and did not differ between inoculated treatments (P > 0.53). In treatment, yeasts had a linear effect in non-inoculated treatment. Mold counting was higher on all days for control treatment and tended to be higher in LPLB supplement on day 6; In days, it had a quadratic effect in control (Table 8). Total gas production was lower in supplements using RCGS (P < 0.01) and similar for inoculated supplements (P = 0.68). It also presented a quadratic behavior as exposure days increase (P = 0.03). Total VFA production differ only between inoculated supplements (P = 0.04), and no differences were found in the other contrasts (P > 0.72) (Table 9). DM tended to differ between non-inoculated and inoculated supplements (P = 0.09) and to have a linear comportment of reduction through days for DM (P < 0.01). No differences in bromatological analyses for supplements were found between inoculated treatments (P > 0.25) but fat (P < 0.01) was higher for inoculated treatments in comparison to non-inoculated and tended to be higher for NDF (P = 0.06). Starch and PB decreased linearly (P < 0.01) and NDF (P = 0.07) tended to decrease linearly while EE had a quadratic effect (P < 0.01) (Table 10). 36 Table 8: Aerobic stability, DM loss, pH and microbial counts of supplements made with RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) Itema Treatment (T) SEM Exposing days (D) SEM P-valueb CON LBLH LBLP 0 3 6 T D T x D C1 C2 L Q AS (hours) 107.5 >144 >144 - - - <0.01 - - <0.01 1.00 - - DM loss (g/kg DM) 32.3 25.7 22.9 0.19 - 22.8 35.4 2.88 0.56 <0.01 0.05 0.40 0.26 <0.01 - pH 6.03 5.68 5.69 0.02 5.26 5.78 6.36 0.02 <0.01 <0.01 <0.01 - - - - Yest (log CFU/g) 5.09 0.18 0.33 0.18 1.69 1.69 2.21 0.25 <0.01 <0.01 <0.01 - - - - Molds (log CFU/g) 1.97 1.94 2.13 0.11 1.77 1.79 2.48 0.11 0.42 <0.01 <0.01 - - - - aAS = aerobic stability, DM loss = dry matter loss during aerobic stability test. bC1: CON vs LBLH + LPLB; C2: LBLH vs LPLB. L: linear contrast effect between exposing days; Q: quadratic contrast effect between days of exposure. Table 9: Total gas and VFA production of supplements made with RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I- 4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) Treatment (T) SEM Exposing days (D) SEM P-valueb Item CON LBLH LBLP 0 3 6 T D T x D C1 C2 L Q Total gas prod (ml/g DM) 198 186 188 3.52 179 198 193 2.71 0.07 0.01 0.06 - - - - Total VFA (Mm/g DM) 31.0 33.0 29.6 1.75 31.1 30.8 31.6 1.63 0.13 0.85 0.22 0.74 0.04 0.99 0.75 bC1: CON vs LBLH + LPLB; C2: LBLH vs LPLB.L: linear contrast effect between exposing days; Q: quadratic contrast effect between days of exposure. 37 Figure 7: Temperature of supplements made with RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) during aerobic stability test. Figure 8: DM loss in supplements made with RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) exposed for 0, 3 and 6 days. Significant contrast effects (C1 and C2) of inoculant over aerobic exposing time: 3 d: NS. 6 d: C1, P < 0.04; C2, NS. Significant contrast effects (3 vs 6) of exposing length over treatment: CON: P < 0.01. LBLH: P = 0.01. LPLB: P = 0.30. *NS = P-value not statistically significant. 22.00 27.00 32.00 37.00 42.00 47.00 0 20 40 60 80 100 120 140 T e m p e ra tu re , °C Hours Room temperature CON LBLH LPLB 0 5 10 15 20 25 30 35 40 45 50 0 3 6 D M l o s s ( g /k g D M ) Days of exposure CON LBLH LBLP 38 Figure 9. pH in supplements made with RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) exposed for 0, 3 and 6 days. Significant contrast effects (C1 and C2) of inoculant over aerobic exposing time: 0 d: C1, P < 0.01; C2 P < 0.01. 3 d: NS. 6 d: C1, P < 0.01; C2, NS. Significant contrast effects (L and Q) of exposing length over treatment: CON: L, P < 0.01; Q, P < 0.01. LBLH: P < 0.01; Q, P < 0.01. LPLB: P < 0.01; Q, P < 0.01. *NS = P-value not statistically significant. Figure 10. Yeast counts in supplements made with RCGS without inoculant (CON) and inoculated L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) exposed for 0, 3 and 6 days. Significant contrast effects (C1 and C2) of 4.50 5.00 5.50 6.00 6.50 7.00 7.50 0 3 6 p H Days of exposure, d CON LBLH LBLP 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 0 3 6 Y a s ts , lo g c fu / g Days of aerobic exposure CON LBLH LBLP 39 inoculant over aerobic exposing time: 0 d: C1, P < 0.01; C2, NS. 3 d: C1, P < 0.01; C2, NS. 6 d: C1, P < 0.01; C2, NS. Significant contrast effects (L and Q) of storage length over treatments: CON: L, P < 0.01; Q, NS. LBLH: NS. LPLB: NS. *NS = P value not statistically significant Figure 11. Molds counts in supplements made with RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) exposed for 0, 3 and 6 days. Significant contrast effects (C1 and C2) of inoculant over aerobic exposing time: 0 d: C1, P < 0.01; C2, NS. 3 d: C1, P < 0.01; C2, NS. 6 d: C1, P < 0.01; C2, NS. Significant contrast effects (L and Q) of storage length over treatments: CON: L, P < 0.01; Q, NS. LBLH: NS. LPLB: NS. *NS = P value not statistically significant. 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 0 3 6 M o ld s , lo g c fu / g Days of exposure CON LBLH LBLP 40 Figure 12: Ash concentration in supplements made with RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) exposed for 0, 3 and 6 days. Significant contrast effects (C1 and C2) of inoculant over aerobic exposing time: 0 d: C1, NS; C2, 0.03. 3 d: C1, NS; C2, NS. 6 d: C1, 0.09; C2, NS. Significant contrast effects (L and Q) of storage length over treatments: CON: L, P = 0.02; Q, NS. LBLH: L, P = 0.09; Q, P= 0.03. LPLB: L, NS; Q, NS. *NS = P value not statistically significant. 40 41 42 43 44 45 46 47 48 49 0 3 6 A s h ( g /k g D M ) Days of exposure CON LBLH LBLP 41 Figure 13: Gas production in supplements made with RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) exposed for 0, 3 and 6 days. Significant contrast effects (C1 and C2) of inoculant over aerobic exposing time: 0 d: C1, NS; C2, NS. 3 d: C1, P < 0.01; C2, NS. 6 d: C1, P < 0.01; C2, NS. Significant contrast effects (L and Q) of storage length over treatments: CON: L, P < 0.01; Q, < 0.01. LBLH: NS. LPLB: L, P = 0.06. 140 150 160 170 180 190 200 210 220 0 3 6 G a s p ro d u c ti o n ( m M o l/ k g D M ) Days of exposure CON LBLH LBLP 42 Table 10: Chemical composition of supplements made with RCGS without inoculant (CON) and inoculated with L. hilgardii CNCM I-4785 (1.5 x 105 cfu/g) + L. buchneri NCIMB 40788 (1.5 x 105 cfu/g) (LBLH) or L. plantarum, MTD/1 (1 x 105 cfu/g) + L. buchneri PJB/1 (2 x 105 cfu/g) (LPLB) exposed for 0, 3 and 6 days. Itema Treatments (T) SEM Exposing days (D) SEM P-valueb CON LBLH LBLP 0 3 6 T D T x D C1 C2 L Q DM (g/kg) 721 719 721 0.15 724 718 712 0.10 0.14 <0.01 0.26 0.09 0.77 <0.01 0.88 CP (g/kg DM) 192 194 194 2.10 199 193 188 2.81 0.84 <0.01 0.70 0.55 0.99 <0.01 0.65 NDF (g/kg DM) 194 205 200 3.92 205 198 196 4.31 0.09 0.14 0.39 0.06 0.25 0.07 0.68 Starch (g/kg DM) 501 521 513 9.31 531 514 491 9.91 0.29 0.02 0.57 0.13 0.52 <0.01 0.77 Fat (g/kg DM) 44.2 46.8 46.4 0.87 43.7 48.9 44.8 1.15 0.09 <0.01 0.92 0.02 0.77 0.02 <0.01 Ash (g/kg DM) 44.3 43.5 44.4 0.72 43.2 44.7 44.3 0.76 0.61 0.29 0.01 0.66 0.38 0.16 0.27 TDN (g/kg DM) 781 778 781 2.42 773 787 780 2.43 0.54 0.01 0.25 0.66 0.31 <0.01 <0.01 a DM = dry mater; CP = crude protein; NDF = neutral detergent fiber; TDN = total digestible nutrients. b C1: CON vs LBLH + LPLB; C2: LBLH vs LPLB; L: linear contrast effect between exposing days; Q: quadratic contrast effect between days of exposure. 43 4. Discussion Higher DM loss in inoculated silage was due to obligatory heterofermentative LAB that produces ethanol and CO2 which leads to higher DM loss (MCDONALD; HENDERSON; HERON, 1991). DM loss in silage during fermentation is mainly from CO2 production (BORREANI et al., 2018). Conversion of lactic acid to acetic acid by L. buchneri also produces CO2 (OUDE ELFERINK et al., 2001) in which increase DM loss during fermentation. The higher LAB number in both inoculated silages than the number of yeasts in CON silage also help to explain the higher DM loss during fermentation, once yeasts produce ethanol and CO2 (DA SILVA et al., 2021). The lower pH in CON silage plus the high number of yeasts at opening indicates the low rate of obligatory heterofermentative LAB and more homofermentative LAB that only produces lactic acid without CO2 losses. The use of L. Plantarum should decrease pH faster due to its metabolism, which avoids secondary fermentation that increases DM loss, however, DM loss data are inconsistent when L. buchneri is added to the silage (DA SILVA et al., 2021; KLEINSCHMIT; KUNG, 2006c), and when acetic acid is produced in significant amount, the inclusion of L. Plantarum is not capable of subtract L. buchneri effects on DM loss (BENJAMIM DA SILVA et al., 2024). Higher values of DM loss in this experiment in comparison with others (DA SILVA et al., 2021, 2024) can be due to enhanced fermentation of finer particle size of grain corn (SAYLOR et al., 2020). It was expected lower pH in LPLB due to the use of Lactobacillus Plantarum as it produces mainly lactic acid (MUCK et al., 2018). Still, Lactobacillus buchneri can degrade lactic acid and produce acetic acid under anaerobic conditions as a mechanism of survival in low pH environments (OUDE ELFERINK et al., 2001) and this might act increasing pH in LPLB silages as concentration of L. buchneri was greater than L. plantarum. As in this experiment, corn grain silages inoculated with LBLH and LPLB had higher pH values than the control silage in other experiments (DA SILVA et al., 2021, 2024). Plant and microbial enzymes can degrade protein in peptides and amino acids. Microorganisms can deaminate amino acids for their use resulting in N-NH3 as a subproduct of their metabolism. Greater N-NH3 concentration in inoculated silages 44 suggests greater proteolysis and it can be due to higher pH once enzymes are deactivated in low pH. Additionally, even though LAB has low protein activity, L. Buchneri may benefit proteolytic microorganisms by consuming lactic acid (JUNGES et al., 2017). The tendency for more N-NH3 in LBLH silage reinforces the idea that L. hilgardii has a similar metabolism and starts its activity earlier. The main source of protein in corn endosperm is in prolamin form covering starch granules and protecting them from ruminal degradation. Increase in N-NH3 is associated with a decrease in prolamin as it is a result of deamination (DA SILVA et al., 2018). Greater MS degradability at 12 hours in inoculated silages is correlated with high degradation of prolamin by bacteria, as they are responsible for 60.4% of proteolytic activity (JACOVACI et al., 2021) and inoculated bacteria seem to be essential in this process as epiphyte LAB population is drastically reduced in the dry processing of corn kernels (Carvalho et al., 2015). L. buchneri inoculation increased DM degradability in other experiments (DA SILVA et al., 2018) and better degradability for LBLH is related to greater N-NH3 that can be a result of early start of action by L. hilgardii using substrate that could be used for other lactic acid producers decreasing the pH and making it difficult for the action of proteolytic microbes. Inoculation with specialized bacteria usually reduces bacteria diversity (DA SILVA et al., 2021, 2024). The low pH in control silage coupled with an anaerobic environment prevented mold growth. In inoculated silage, even though the higher pH, acetic acid avoided molds and yeast growth. The absence of acetic acid and the low pH tolerance favored yeast growth in the control silage. The greater concentration of obligatory heterofermentative bacteria in LBLH inoculant promoted greater production of acetic acid. Greater concentration of 1,2 propanediol in inoculated silages indicates the activity of L. buchneri. Higher concentration of this bacteria in LPLB led to a higher concentration of 1,2 propanediol when in relation to LBLH. The very low concentration of butyric acid indicates good fermentation during storage period. Higher acetic ac